Skip to main content

8 AI chatbots you should use instead of ChatGPT

Copilot on a laptop on a desk.
Microsoft

When ChatGPT launched in late 2022, it was a novelty. It didn’t take long, however, for competition to come along.

Early on, there weren’t many ChatGPT alternatives available that weren’t in-house, research-based options or open source projects on GitHub that required some sort of coding knowledge to set up and operate. But since then, several companies have developed consumer products with free and paid tiers and a plethora of enterprise and developer options. So, if you aren’t satisfied with ChatGPT for whatever reason, these are the eight other options to try out instead.

Microsoft Copilot

Windows 11 Copilot in Windows 11.
Jon Martindale / DigitalTrends

Microsoft Copilot has come a long way since its first chatbot iteration was released in early 2023. It was one of ChatGPT’s first commercial competitors when the brand introduced Bing AI Chat as an integration into its Bing Search engine and Edge browser. Since then, the company has rebranded its chatbot to Microsoft Copilot in its browser and also integrated its functionality across several services, including the Microsoft 365 productivity suite, Microsoft Designer, and the Windows operating system, among others. Microsoft also offers a paid Copilot Pro option for power users.

Notably, Copilot is powered by OpenAI’s GPT-4 large language model (LLM) and DALL-E image generation model due to Microsoft’s investment in the company. The brand includes a mix of its own proprietary technology within the chatbot, which differentiates its results from ChatGPT. Some consider its text results to be stronger due to being more directly connected to the internet, but claim that OpenAI produces stronger images natively with its models.

Google Gemini

Google Gemini Theory of Relativity.
Google

Google Gemini is another chatbot that has seen several versions, rebranding, and tiered options since its initial introduction. Google’s first AI endeavor was a research offering called Bard, which ran on Google’s LaMDA LLM and debuted in March 2023. Not long after, the brand made Google Duet available as an AI-inundated enterprise option for its Workspace apps, including Gmail, Drive, Slides, Docs, and others. By December 2023, Google upgraded the Bard language model to the Gemini LLM. In February, the company merged the Duet and Bard services into a single product, branding all of its AI options under the name Gemini.

Google now supports several LLM options that are available for different tiers of purpose and expertise, including Gemini Nano, Gemini Pro, Gemini Ultra, and Gemini Advanced. Gemini Pro powers the free Google Gemini chatbot that is the brand’s entry-level,web- and app-based AI product. Gemini Advanced powers the paid Google One AI Premium tier. The company recently released a public preview of the Gemini 1.5 Pro LLM that enables “hearing” capabilities when audio files, from which the chatbot can extract the text information, are uploaded to a system.

Gemini is a direct competitor for both ChatGPT and Microsoft Copilot.

Perplexity

Someone holding an iPhone 14 Pro, with Perplexity AI running on it.
Joe Maring / Digital Trends

Perplexity AI is a research-focused chatbot with a company CEO who is an OpenAI alumnus. Its simple user interface is reminiscent of ChatGPT, but it doesn’t require an initial login to test. You can simply begin typing your query or click one of the suggested topics available to begin a conversation. Clicking the discover tab also brings up recent news similar to a search engine. The AI chatbot is known for citing its sources with links and being up to date with information pulled from the internet.

Perplexity’s access includes multiple tiers. The first aforementioned iier allows you to generate conversations anonymously without signing up. Then there is the free sign-up tier, which includes a single-use sign-up option, and a paid Perplexity Pro tier. Like many chatbots, the free and paid tiers use different language models. The free tiers are based on OpenAI’s GPT-3.5 model and also use the natural language processing (NLP) model to train the chatbot. Meanwhile, Perplexity Pro uses a mix of GPT-4, Claude 3, Mistral Large, Llama 3 and an Experimental Perplexity model for different processes.

In addition to ChatGPT, Microsoft Copilot is considered one of Perplexity’s closest competitors.

Chatsonic

A screenshot of the Chatsonic AI tools page.
Chatsonic

Chatsonic is an interesting ChatGPT alternative offered by the Writesonic brand for writing assistance and content creation. The brand offers several free extensions that work in the Google Chrome browser, ChatGPT web version, Gmail, and X (formerly Twitter). Its sign-up options begin with a free tier with a one-time offer of 25 credits to try out the Chatsonic and Writesonic services. There are pricing tiers for individuals and freelancers, as well as tiers for professionals and teams. Similar to other brands, Chatsonic is currently based on the GPT-3.5 LLM for its free version, GPT-4 LLM and Claude 3 Opus for its paid tiers, and the DALL-E and Stable Diffusion image generation models.

The service leans in favor of paid subscriptions, which includes many perks, such as a plagiarism checker and the ability to input documents, images, audio files, and URLs into the chat. The service runs on a token system, so if you run out of tokens for the month, its functionality will be greatly reduced.

Character.AI

The Character AI home page.
Digital Trends

Character.AI is a chatbot service that became very popular in early 2023 when ChatGPT was suffering from capacity issues. Users flocked to the service with the interesting model for a quick respite while OpenAI sorted out its bandwidth. The company is run by former developers of Google’s LaMDA LLM and Meena chatbot, however, the team does not make clear whether Character.AI is based on these technologies.

Nevertheless, the premise of the service is simple, as you can communicate one-on-one with characters that may be based on notable people, fictional characters from books, video games, TV shows or movies, or a conceptual person like a teacher, therapist or coach. Similarly, you can create and train a character, giving them a humanlike personality and introduce them to the Character.AI community. Other features allow you to share your conversations publicly within the community and allow the fictional characters to communicate with each other.

You can test the service for free by beginning a conversation with no sign-up; however, it will eventually prompt you to register after a few turns. Its paid subscription option includes faster response times, early access to new features, and a c.ai+ supporter badge, among other perks.

Anthropic Claude

A screenshot of the Anthropic Claude AI chatbot.
Image used with permission by copyright holder

Claude is a family of LLMs developed by Anthropic and first released in early 2023. IT is known to excel in coding, math, writing, research, and image analysis. Its current iteration, Claude 3, was released in March 2024.

Claude was developed on a “Constitutional AI” training method intended to keep human interaction out of the learning process of the model as much as possible. This helps it to self-correct, which allows it to maintain the safety and integrity of the model.

Despite being on the market for a short time, Claude has proven a popular competitor  of many of the top chatbots currently available, with many brands using the model to power their own tools. Claude also includes features that others may lack, such as the ability to upload and analyze documents, as well as information training up to early 2023.

The chatbot remains in a free beta phase for its entry tier, but there’s also a monthly paid tier.

Pi AI

Discussing nostalgia with Pi assistant on Android.
Nadeem Sarwar / Digital Trends

Pi AI is a unique chatbot that features a simple user interface with no additional features. It is known for its conversation response that provides a high level of emotional engagement for users. Pi AI is developed in a way that it can detect the need for kindness, empathy, diplomacy, or humor in a conversation and maintain the inflections as long as they are needed. Unlike many other chatbots, which emulate a character in a manner that focuses on the character’s personality or collects generic information and regurgitates a script-like dialogue, Pi AI can offer a more humanlike experience by being able to detect the various emotions coming up in queries that it is presented with.

The chatbot was developed by the AI startup Inflection, which has a team made up of AI experts who formerly worked at companies including DeepMind, Google, OpenAI, and Meta.

Poe by Quora

Quora's Claude+ bot not knowing itself.
Quora

Quora, already known as a question-and-answer database, has entered the AI space with an option called Poe, which is a collection of the most well-known LLMs set up in one platform.

Selecting your LLM of choice will load its corresponding chatbot and you can begin a conversation. Some of the models included in Poe are Google PaLM and Gemini, Meta Llama, and Anthropic Claude, multiple versions of the GPT LLMs, and lesser-known technologies including Sage, Dragonfly, and NeevaAI.

Poe by Quora is available for free, but with restricted turns on conversations and other limitations on features. Opting for the monthly subscription will vastly increase the capability of each model and chatbot offered.

Fionna Agomuoh
Fionna Agomuoh is a Computing Writer at Digital Trends. She covers a range of topics in the computing space, including…
ChatGPT’s resource demands are getting out of control
a server

It's no secret that the growth of generative AI has demanded ever increasing amounts of water and electricity, but a new study from The Washington Post and researchers from University of California, Riverside shows just how many resources OpenAI's chatbot needs in order to perform even its most basic functions.

In terms of water usage, the amount needed for ChatGPT to write a 100-word email depends on the state and the user's proximity to OpenAI's nearest data center. The less prevalent water is in a given region, and the less expensive electricity is, the more likely the data center is to rely on electrically powered air conditioning units instead. In Texas, for example, the chatbot only consumes an estimated 235 milliliters needed to generate one 100-word email. That same email drafted in Washington, on the other hand, would require 1,408 milliliters (nearly a liter and a half) per email.

Read more
YouTube can now generate a 6-second video Short using AI
The Digital Trends YouTube channel as seen on an iPhone and on a TV in the background.

YouTube content creators will soon have a slew of new AI-empowered tools at their disposal, the company announced Wednesday at its Made on YouTube event in New York City, including the ability to generate a complete six-second YouTube Shorts video clip with a text prompt.

The new capability arrives thanks to the integration of Google Deep Mind's Veo video generation model into YouTube Shorts. Built to compete with the likes of OpenAI's Sora, Runway's Gen-3 Alpha, Adobe's Firefly or Kuaishou Technology's Kling, Veo can generate six-second clips at 1080p resolution across a wide range of cinematic themes and styles.

Read more
OpenAI’s advanced ‘Project Strawberry’ model has finally arrived
chatGPT on a phone on an encyclopedia

After months of speculation and anticipation, OpenAI has released the production version of its advanced reasoning model, Project Strawberry, which has been renamed "o1." It is joined by a "mini" version (just as GPT-4o was) that will offer faster and more responsive interactions at the expense of leveraging a larger knowledge base.

It appears that o1 offers a mixed bag of technical advancements. It's the first in OpenAI's line of reasoning models designed to use humanlike deduction to answer complex questions on subjects -- including science, coding, and math -- faster than humans can.

Read more